Resting state functional connectivity of the hippocampus associated with neurocognitive function in left temporal lobe epilepsy

Abstract

The majority of patients with temporal lobe epilepsy (TLE) experience disturbances of episodic memory from structural damage or dysfunction of the hippocampus. The objective of this study was to use functional Magnetic Resonance Imaging (fMRI) to identify regions where resting state connectivity to the left hippocampus (LH) is correlated with neuropsychological measures of verbal memory retention in TLE patients. Eleven left TLE (LTLE) patients and 15 control subjects participated in resting state fMRI scans. All LTLE patients underwent neuropsychological testing. Resting state functional connectivity maps to the LH were calculated for each patient, and subsequently used in a multiple regression analysis with verbal memory retention scores as a covariate. The analysis identified brain regions whose connectivity to the LH was linearly related to memory retention scores across the group of patients. In LTLE patients, right sided (contralateral) clusters in the precuneus and inferior parietal lobule (IPL) exhibited increased connectivity to the LH with increased memory retention score; left sided (ipsilateral) regions in the precuneus and IPL showed increased connectivity to the LH with decreased retention score. Patients with high memory retention scores had greater connectivity between the LH-right parietal clusters than between the LH-left parietal clusters; in contrast, control subjects had significantly and consistently greater LH-left hemisphere than LH-right hemisphere connectivity. Our results suggest that increased connectivity in contralateral hippocampal functional pathways within the episodic verbal memory network represents a strengthening of alternative pathways in LTLE patients with strong verbal memory retention abilities.

Keywords: brain; connectivity; episodic memory; functional MRI; resting state; temporal lobe epilepsy.

Copyright © 2012 Wiley Periodicals, Inc.

Figures

Figure 1

Resting state functional connectivity maps to the left hippocampus. (A) Control subject used in the study. (B) LTLE patient in the study.

Figure 2

Regions showing significant correlation between resting state connectivity to the LH and CVLT‐II percentage retention score across group of 11 LTLE patients. (A) Cluster in the mid‐right precuneus (magenta), right inferior parietal lobule (green), and right insula (yellow) showing positive correlation (decreased connectivity with decreased score). (B) Cluster in the left precuneus (yellow), left inferior parietal lobule (red), and left middle frontal gyrus (cyan) demonstrating a negative correlation (increased connectivity with decreased score.

Figure 3

Positive linear correlation between connectivity to the LH and the CVLT‐II percentage retention scores in three brain regions. (A) The right inferior parietal lobule. (B) The mid‐right precuneus. (C) The right insula/precentral gyrus. Patients with MTS are shown in red, and patients without MTS are shown in blue as identified in Table 1. CVLT‐II scores are unitless and beta values have arbitrary units. Linear regression was used to fit a line to the LTLE patient data for each plot. Vertical line represents mean normal CVLT‐II percentage retention score of 0.868 for all ages and genders. See Table 2 for a statistical summary of these regions.

Figure 4

Negative linear correlation between connectivity to the LH and CVLT‐II percentage retention scores in three brain regions. (A) The left inferior parietal lobule. (B) The left precuneus/cingulate gyrus. (C) The left middle frontal gyrus. Patients with MTS are shown in red, and patients without MTS are shown in blue as identified in Table 1. CVLT‐II scores are unitless and beta values have arbitrary units. Linear regression was used to fit a line to the LTLE patient data for each plot. Vertical line represents mean normal CVLT‐II percentage retention score of 0.868 for all ages and genders. See Table 2 for statistical summary of these regions.

Figure 5

(A) Plot of the connectivity between the LH and the right inferior parietal lobule vs. connectivity between the LH and the left inferior parietal lobule for LTLE patients. (B) Plot of the connectivity between the LH and the mid‐right precuneus versus connectivity between the LH and the left precuneus for LTLE patients. We used linear regression to fit a line to the LTLE patient data. The patients were divided into three categories based on their CVLT‐II percentage retention score.